BASALT documentation

BASALT documentation#

Genome-resolved metagenomics · PKU-EMBL Laboratory

Recover and refine metagenome-assembled genomes across one or more assemblies in a checkpointed workflow.

BASALT connects multi-binner candidate generation, within- and between-assembly bin selection, contig-level contamination screening, read-supported retrieval, and reassembly.

Get started Discover BASALT-Air Plan a study Explore the method

Nature Communications 2024 New: BASALT-Air v1.0.0Multi-assemblyMIT licensed

New implementation: BASALT-Air v1.0.0

BASALT-Air is now available for new deployments. It uses Pixi, accepts absolute input paths, separates working and output directories, and records a structured run manifest. Read Choosing BASALT-Air before switching: the executable, sample delimiters, path options, and checkpoints are not interchangeable with this Conda edition.

Evidence boundary

The published study reports improved MAG recovery and downstream genome-resolved analyses relative to the workflows and datasets tested. This is not a guarantee for every community or sequencing design. Dataset complexity, sequencing depth, assembly quality, read type, database choice, and parameter settings can change the outcome.

Start here#

Choose the route that matches the decision you need to make. Commands and defaults on this site are checked against the current repository.

Install and verify

Configure BASALT, model weights, quality-control databases, external executables, and containers.

Installation

Run a minimal analysis

Launch an isolated short-read run with explicit resources, mode, and output name.

Quick start

Design assemblies and coverage

Choose between sample-specific, pooled, longitudinal, and multi-assembly strategies.

Study design patterns

Understand the method

Follow each executable stage and distinguish software behaviour from interpretation.

Pipeline and methods

Audit the outputs

Locate final MAGs, inspect quality reports, and diagnose incomplete stages.

Outputs and quality control

Report the analysis

Archive provenance and use the minimum Methods and reporting checklist.

Reproducibility and reporting

BASALT framework#

Method at a glance#

Stage	Function	Main evidence used	Principal boundary
Autobinning	Generate candidate bins under several algorithms or parameter settings	Sequence composition and read depth	Candidate diversity depends on the selected preset and optional binners
Bin selection	Compare bins within and across assemblies	Bin quality estimates, overlap, composition, and coverage	Dereplication does not establish biological strain identity
Refinement	Screen candidate contaminant contigs and retrieve supported contigs	Composition, coverage-derived features, and read connectivity	Predictions depend on the trained models and available coverage signals
Reassembly	Reassemble selected bins when compatible reads are present	Reads assigned to individual bins	Reassembly can be skipped or fail for bins with insufficient read support

The pipeline page separates what the software executes from how results should be interpreted.

Supported analysis designs#

Every run requires one or more assembly FASTA files and at least one read source that can provide coverage information.

Design	Short reads	ONT or PacBio CLR	PacBio HiFi
Short-read only	required	no	no
Long-read only	no	required	no
HiFi only	no	no	required
Short reads + ONT or CLR	required	required	no
Short reads + HiFi	required	no	required

Read Input formats and Study design patterns before combining assemblies or samples. BASALT does not infer which reads belong to which biological sample.

BASALT and BASALT-Air#

This site primarily documents the Conda-based BASALT repository. BASALT-Air v1.0.0 is the newer Pixi-based implementation with absolute-path handling, separate work and output directories, dependency checks, and structured run manifests. See Choosing BASALT-Air for a concise comparison.

Use the documentation that belongs to the implementation and version you ran. Do not mix BASALT-Air separators or path options into a BASALT command.

Evidence and citation#

If BASALT contributes to an analysis, cite:

Qiu, Z., Yuan, L., Lian, C.-A. et al. BASALT refines binning from metagenomic data and increases resolution of genome-resolved metagenomic analysis. Nature Communications 15, 2179 (2024). https://doi.org/10.1038/s41467-024-46539-7

@article{qiu2024basalt,
  author  = {Qiu, Zhiguang and Yuan, Li and Lian, Chun-Ang and Lin, Bin and
             Chen, Jie and Mu, Rong and Qiao, Xuejiao and Zhang, Liyu and
             Xu, Zheng and Fan, Lu and Zhang, Yunzeng and Wang, Shanquan and
             Li, Junyi and Cao, Huiluo and Li, Bing and Chen, Baowei and
             Song, Chi and Liu, Yongxin and Shi, Lili and Tian, Yonghong and
             Ni, Jinren and Zhang, Tong and Zhou, Jizhong and Zhuang, Wei-Qin and
             Yu, Ke},
  title   = {BASALT refines binning from metagenomic data and increases resolution of genome-resolved metagenomic analysis},
  journal = {Nature Communications},
  volume  = {15},
  pages   = {2179},
  year    = {2024},
  doi     = {10.1038/s41467-024-46539-7}
}

Also cite optional binners and quality-control software when they materially affect the reported result.

Scope of this documentation#

The guide distinguishes three kinds of statement:

Executable behaviour, checked against the current repository.
Scientific interpretation, limited to what the workflow and published evidence support.
Operational guidance, identified as a recommendation when it is not a program requirement.

If documentation and observed behaviour differ, retain the full command, Basalt_log.txt, Basalt_checkpoint.txt, software versions, and a minimal reproducer. Report the discrepancy through GitHub Issues.

BASALT documentation

Contents

BASALT documentation#

Core advantage: coherent multi-assembly refinement#